Silent <k> and <g> in English
-- UKT 120212
Pronouncing the velar-plosive-stops /k/ and /g/ before the dental sounds in Modern English is a problem. I have found a similar situation in the pronunciations of Sanskrit [क ् ष = क्ष] and Pali [चक्-ख].
[The following presentation needs Myanmar graphemes, however, since they do not appear in emails, I have given their equivalents in Devanagari. Also note that Bur-Myan {sa.} stands for both palatal plosive-stop [च] /c/ and dental fricative-sibilant [ष] /s/. However, they are differentiated in Romabama as {sa.} for [च] and {Sa.} for [ष]. ]
In {sa.kSa.} [च-क ् ष], we have two syllables, {sa.} [च] and {kSa.} [क ् ष = क्ष] if you are using the Sanskrit pronunciation. The conjunct {kSa.} [क ् ष = क्ष ] is not readily pronounceable, because {ka.} [क] /k/ the velar has to be pronounced immediately before the dental {Sa.} [ष] /s/. The pronunciation I can come up is /kə.sa/ with a schwa after /k/. Thus, {kSa.} [क ् ष = क्ष ] cannot become a medial but remains a disyllabic conjunct.
This reminds me of cases of a velar-before-dental in English <know> and <gnome>. In modern English both <k> and <g> are made 'silent'. Such silent letters are not permissible in the akshara-system which is a very rigid phonetic system preceding the International Phonetic Association (IPA) by thousands of years.
Since क्ष «ks» is not readily pronounceable for Bur-Myan and presumably for Pal-Myan, it is pronounced as {hka.} [ख] . Thus I have dubbed as pseudo-{hka.}.
An interesting point is what happens when the two syllables are pronounced very rapidly as in continuous speech. This amounts to the graphemes moving closer. {kSa.} conjunct in {sa~kSa.} is broken apart, and the pronunciation becomes: {sak~Sa.} [चक्-ष] . Since dental fricative-sibilant {Sa.} [ष] is absent in Bur-Myan, it is changed into {hka.} [ख] : resulting {sak~hka.} [चक्-ख]. I am waiting for input from my peers.

